Air pollutants arising from traffic clustering and industrial production include a number of chemical compounds that at high doses are carcinogenic in animal models and in some instances also in humans. Direct epidemiological evidence for a carcinogenic effect of air pollution in humans is, however, weak, and most of the available studies are limited by lack of adequate control of confounding factors and other methodological drawbacks. Limited evidence exists for a link between urban air pollution and lung cancer, with reported relative risks of 1.0-1.5. About one-third of the population of the Nordic countries, corresponding to 7.3 million people, lives in urban areas. If there is an excess risk associated with air pollution, the annual number of lung cancer cases around the year 2000 in the Nordic countries would range from 0 (no excess risk) to 1,800 (relative risk, 1.5). As the existence of a causal link between air pollution and cancer is not uncorroborated, measures for avoiding cancer from this source cannot be recommended.